CentOS7单节点和多节点HPL测试

前置工作：安装OpenBLAS; 安装Mpich (可参考首页博客)

• 官网下载压缩包到/opt目录

  cd /opt && wget https://www.netlib.org/benchmark/hpl/hpl-2.3.tar.gz

  

• 解压到 /opt 目录

  tar -xzf hpl-2.3.tar.gz

• 复制Make.Linux_PII_CBLAS并重命名

  cd /opt/hpl-2.3 && cp setup/Make.Linux_PII_CBLAS Make.Linux

• 编辑Make.Linux

  vim Make.Linux

  修改如下内容：

  ARCH = LinuxTOPdir = /opt/hpl-2.3  # hpl安装目录MPdir = /opt/mpich     # mpich安装目录MPlib = $(MPdir)/lib/libmpi.a # mpi链接库LAdir = /opt/OpenBLAS # openblas安装目录LAlib = $(LAdir)/lib/libopenblas.a  # openblas链接库CC = /opt/mpich/bin/mpicc # compilerCCFLAGS = $(HPL_DEFS) -fomit-frame-pointer -O3 -funroll-loops -pthreadLINKER = /opt/mpich/bin/mpif77 # linker

  以上路径根据个人安装时的目录修改

• 构建hpl

  make arch=Linux

  若build成功，则会在/opt/hpl-2.3/bin/Linux目录下生成HPL.dat和xhpl文件

• 测试hpl

  cd /opt/hpl-2.3/bin/Linux

  1. 单节点测试

     mpiexec -np 4 ./xhpl

  2. 多节点测试

     编辑节点文件，输入节点主机名或IP地址

     vim nodes

     eg:

     

     修改HPL.dat

     HPLinpack benchmark input fileInnovative Computing Laboratory, University of TennesseeHPL.out      output file name (if any)6            device out (6=stdout,7=stderr,file)1            # of problems sizes (N)1200         Ns1            # of NBs232          NBs0            PMAP process mapping (0=Row-,1=Column-major)1            # of process grids (P x Q)1            Ps4            Qs16.0         threshold1            # of panel fact0            PFACTs (0=left, 1=Crout, 2=Right)1            # of recursive stopping criterium2            NBMINs (>= 1)1            # of panels in recursion2            NDIVs1            # of recursive panel fact.0            RFACTs (0=left, 1=Crout, 2=Right)1            # of broadcast0            BCASTs (0=1rg,1=1rM,2=2rg,3=2rM,4=Lng,5=LnM)1            # of lookahead depth1            DEPTHs (>=0)2            SWAP (0=bin-exch,1=long,2=mix)64           swapping threshold0            L1 in (0=transposed,1=no-transposed) form0            U  in (0=transposed,1=no-transposed) form1            Equilibration (0=no,1=yes)8            memory alignment in double (> 0)

     运行hpl

     mpiexec -np 4 -machinefile ./nodes ./xhpl

     

  3. HPL.dat配置项解释

     HPLinpack benchmark input file                            # 文件头，说明Innovative Computing Laboratory, University of TennesseeHPL.out      output file name (if any)                 # 如果使用文件保留输出结果，设定文件名6            device out (6=stdout,7=stderr,file)     # 输出方式选择（stdout,stderr或文件）2            # of problems sizes (N)              # 指出要计算的矩阵规格有几种1960  2048   Ns                                           # 每种规格分别的数值2            # of NBs                             # 指出使用几种不同的分块大小60 80        NBs                                     # 分别指出每种大小的具体值2            # of process grids (P x Q-l         # 指出用几种进程组合方式2   4        Ps                                  # 每对PQ具体的值2   1        Qs                                   16.0         threshold                           # 余数的阈值1            # of panel fact                     # 用几种分解方法1            PFACTs (0=left, 1=Crout, 2=Right)    # 具体用哪种,0 left,1 crout,2 right1            # of recursive stopping criterium    # 几种停止递归的判断标准4            NBMINs (>= 1)                         # 具体的标准数值（须不小于1）1            # of panels in recursion              # 递归中用几种分割法2            NDIVs                               # 这里用一种NDIV值为2，即每次递归分成两块1            # of recursive panel fact.          # 用几种递归分解方法2            RFACTs (0=left, 1=Crout, 2=Right)      # 这里每种都用到（左，右，crout分解）1            # of broadcast                          # 用几种广播方法3            BCASTs (0=1rg,1=1rM,2=2rg,3=2rM,4=Lng,5=LnM)   # 指定具体哪种（有1-ring,1-ring Modified,2-ring,2ring Modified,Long以及long-Modified）1            # of lookahead depth     # 用几种向前看的步数1            DEPTHs (>=0)             # 具体步数值（须大于等于0）2            SWAP (0=bin-exch,1=long,2=mix)  # 哪种交换算法（bin-exchange，long或者二者混合）64           swapping threshold     # 采用混合的交换算法时使用的阈值0            L1 in (0=transposed,1=no-transposed) form     # L1是否用转置形式0	U  in (0=transposed,1=no-transposed) form    # U是否用转置形式表示1            Equilibration (0=no,1=yes)                # 是否采用平衡状态8            memory alignment in double (> 0)     # 指出程序运行时内存分配中的采用的对齐方式